Guardado en:
Detalles Bibliográficos
Autores principales: Zhong, Hanyang, Wang, Liman, Cao, Wenting, Sun, Zeyuan
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2406.10999
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915238964625408
author Zhong, Hanyang
Wang, Liman
Cao, Wenting
Sun, Zeyuan
author_facet Zhong, Hanyang
Wang, Liman
Cao, Wenting
Sun, Zeyuan
contents This paper examines the role of cognitive biases in the decision-making processes of large language models (LLMs), challenging the conventional goal of eliminating all biases. When properly balanced, we show that certain cognitive biases can enhance decision-making efficiency through rational deviations and heuristic shortcuts. By introducing heuristic moderation and an abstention option, which allows LLMs to withhold responses when uncertain, we reduce error rates, improve decision accuracy, and optimize decision rates. Using the Balance Rigor and Utility (BRU) dataset, developed through expert collaboration, our findings demonstrate that targeted inspection of cognitive biases aligns LLM decisions more closely with human reasoning, enhancing reliability and suggesting strategies for future improvements. This approach offers a novel way to leverage cognitive biases to improve the practical utility of LLMs across various applications.
format Preprint
id arxiv_https___arxiv_org_abs_2406_10999
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Balancing Rigor and Utility: Mitigating Cognitive Biases in Large Language Models for Multiple-Choice Questions
Zhong, Hanyang
Wang, Liman
Cao, Wenting
Sun, Zeyuan
Computation and Language
Artificial Intelligence
This paper examines the role of cognitive biases in the decision-making processes of large language models (LLMs), challenging the conventional goal of eliminating all biases. When properly balanced, we show that certain cognitive biases can enhance decision-making efficiency through rational deviations and heuristic shortcuts. By introducing heuristic moderation and an abstention option, which allows LLMs to withhold responses when uncertain, we reduce error rates, improve decision accuracy, and optimize decision rates. Using the Balance Rigor and Utility (BRU) dataset, developed through expert collaboration, our findings demonstrate that targeted inspection of cognitive biases aligns LLM decisions more closely with human reasoning, enhancing reliability and suggesting strategies for future improvements. This approach offers a novel way to leverage cognitive biases to improve the practical utility of LLMs across various applications.
title Balancing Rigor and Utility: Mitigating Cognitive Biases in Large Language Models for Multiple-Choice Questions
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2406.10999